Linux-Misc Digest #782, Volume #24 Mon, 12 Jun 00 02:13:02 EDT
Contents:
Re: Drivers ("Java__Cat")
Re: vote on MS split-up (R.E.Ballard ( Rex Ballard ))
What should I do with ISO file? (Paul)
Re: Sun Sparc faster then intel pentium: is this true???? (Stephen E. Halpin)
Security ("David ..")
Re: What should I do with ISO file? (Jehsom)
Re: Help - Questions... (Jehsom)
Re: Sun Sparc faster then intel pentium: is this true???? (Stephen E. Halpin)
point to point
----------------------------------------------------------------------------
From: "Java__Cat" <[EMAIL PROTECTED]>
Subject: Re: Drivers
Date: Mon, 12 Jun 2000 04:20:10 GMT
Well still have not figured it out. I hate this darn 640x480 resolution. The
system just will not let me adjust the size of the resolution. I am using
SuSE and it comes with a utillity called SAX and that does not seem to like
me.
"David M. Cook" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> On Mon, 12 Jun 2000 02:43:47 GMT, Java__Cat <[EMAIL PROTECTED]>
wrote:
>
> >I am a Linux newbie in need of drivers. Does any one know of a good site
to
> >locate them, heck I'll even take a bad site if it has the right drivers.
The
> >most important one is for my Monitor I have an HP D3859A MultiMedia 17in.
> >display.
>
> Monitors don't have drivers per se. The behavior of the display is
handled
> through "modelines" in /etc/X11/XF86Config. Most monitors work with
> standard VESA modes and you shouldn't need to mess with modelines.
>
> Are you having a problem configuring X?
>
> Dave Cook
------------------------------
From: R.E.Ballard ( Rex Ballard ) <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.advocacy
Subject: Re: vote on MS split-up
Date: Mon, 12 Jun 2000 04:07:16 GMT
In article <[EMAIL PROTECTED]>,
Smitty <[EMAIL PROTECTED]> wrote:
> "Colin R. Day" wrote:
>
> > Gerald Willmann wrote:
> >
> > > CNN is conducting a poll whether MS should be split up and if yes
into how
> > > many parts. Please take a minute to vote for a good cause.
> > >
> > > -> http://cnnfn.com/poll/microsoft_breakup.html
> > >
> > > thanks, Gerald
> > >
> > > --
> >
> > Justice is not the product of opinion polls. Besides, I want
> > Microsoft destroyed by Linux, not the DOJ.
There's only one problem. Microsoft is playing with a loaded deck,
marked cards, cameras behind each of the players, and all of the
aces up their sleeves. And they've slipped mikkies in the drinks.
Moving away from the metaphores:
Microsoft has binding contracts with the top 20 OEMs that prevent
them from making any changes that would allow them to install Linux
on the same computer. Furthermore, Microsoft uses cliff tiered pricing
to force them to buy 120% of their capacity. Finally, Microsoft has
already threatened license audits for machines which did not have
Windows installed at shipment. This is what they did to IBM, and
they claimed that they were justified in doing it to protect themselves
from piracy.
Although the Judge has already judged these contracts to be illegal and
therefore null and void effective immediately, the industry is acting as
if they were valid for the next 90 days for fear that Microsoft will win
a stay of the entire verdict (meaning that the contracts could be
legally binding retroactively).
That's the loaded deck.
Second, Microsft has repeatedly exploited code developed under the BSD
license, the NCSA public license, and the other public license variants
that don't explicitly state that all enhancements must be submitted
back to the original author for possible inclusion into the standard
product. Keep in mind that Internet Explorer was based on NCSA Mosaic
which was sold to Microsoft by Spyglass under terms never formally
ratified by the developers.
That's the marked cards.
Because Linux is based on Open Source, specifically the GNU Public
License, and because it is based on published standards, there are
very few secrets. Microsoft knows what we are doing, how we are
doing it, how we operate, how we develop, and how we market.
This is the camera behind every player.
Microsoft has attempted to isolate the Linux community through a
series of nondisclosure agreements with all of the standards
committees which were a prerequisite of Microsoft's participation.
While the significance of these NDAs was minimized as a "standard
formality of no significance", Microsoft has shown it's true
intent in it's threat to sue Slash/Dot for not removing a posted
copy of the Kerberos 5 specification preceeded with a Microsoft
nondisclosure agreement. The same document was posted on Usenet.
The fact is that Microsoft has used their NDAs to exclude Linux
from PCI PnP, USB, DVDCSS, and has attempted to use it to exclude
Linux from MSCHAP logins to AOL, MCI, and Sprint dial-up connections.
They also attempted to use NDAs to protect NTFS, and are using it
to protect NTFS2. They have even been trying to protect LDAP
and CORBA!
That's the aces up the sleeves.
Microsoft has also been using their proprietary trade secret code
to commit various forms of electronic vandalism aimed at causing
unexpainable crashes of competitor products such as Netscape,
Lotus Notes, and WordPerfect. This wouldn't be the first time
(MS-DOS 6.0 and Stacker, NT 4.0 SP2 and Cyrix, NT 4.0 SP 4 and ???).
Microsoft's latest dirty tricks include "Medialess" Windows 2000
licenses guaranteed to force corporate users to avoid Linux at all
costs for fear of losing the ability to recover the hard drive
after a crash. This could be reasonable and customary if the media
is available for $10 to $15.
That's the Mikkies in the drinks.
Essentially, the "interim" remedies would be sufficent to level the
playing field a little. Only half the cards would be marked.
Only have the players would have cameras and we'd know which
ones, and only two of the aces would be left up the sleeve.
And the drinks will be too strong, but chemical free.
The interim remedies is more like Probation. If Microsoft could
play by the rules until the appeals are complete, they have a
very good chance of beating the structural remedy. On the other
hand, if they keep jerking around, acting as if the entire verdict
was a joke and that they are above the law, it will give the
prosecutors an overwhelming argument that proves that the structural
remedy is necessary.
> > Colin Day
> Hear! Hear! Crush them in the market place where they live!
But lets not bring knives to a gunfight.
> Smitty
--
Rex Ballard - Open Source Advocate, Internet
I/T Architect, MIS Director
http://www.open4success.com
Linux - 90 million satisfied users worldwide
and growing at over 5%/month!
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: Paul <[EMAIL PROTECTED]>
Subject: What should I do with ISO file?
Date: Mon, 12 Jun 2000 04:31:01 GMT
Hi! I download Redhat 6.2, but it is in iso files. How can I convert it to a normal
files so that I can install it and also be able to copy it to a cd-r? Please explain
more detail for me because I am a beginner. Thank you very much for your
help. I am really appreciated.
Paul
--
Posted via CNET Help.com
http://www.help.com/
------------------------------
From: [EMAIL PROTECTED] (Stephen E. Halpin)
Crossposted-To: comp.sys.sun.hardware,comp.sys.sun.misc
Subject: Re: Sun Sparc faster then intel pentium: is this true????
Date: Mon, 12 Jun 2000 04:45:14 GMT
On Sat, 10 Jun 2000 15:47:58 -0500, "Andrew N. McGuire " <[EMAIL PROTECTED]>
wrote:
>On Sat, 10 Jun 2000, Stephen E. Halpin wrote:
>
>+ On Fri, 9 Jun 2000 20:51:43 -0500, "Andrew N. McGuire " <[EMAIL PROTECTED]>
>wrote:
>+
>+ >On Sat, 10 Jun 2000, Stephen E. Halpin wrote:
>+ >
>+ >+ On Fri, 09 Jun 2000 16:35:40 GMT, Rich Teer <[EMAIL PROTECTED]> wrote:
>+ >+
>+ >+ >On Fri, 9 Jun 2000, Stephen E. Halpin wrote:
>+ >+ >
>+ >+ >> On Wed, 07 Jun 2000 04:23:07 GMT, "Toaster Tester" <[EMAIL PROTECTED]> wrote:
>+ >+ >>
>+ >+ >> >I have read a few comparision of Intel vs. RISC processor and they tend to
>+ >+ >> >come up the same. For integer instructions on similar MHZ CPUs Intel and
>+ >+ >> >RISC are usually even. Where Intel loses big is floating point
>+ >+ >> >instructions, RISC runs circles around Intel on floating point.
>+ >+ >>
>+ >+ >> It depends on what you are doing, and which RISC. The shipping 300MHz
>+ >+ >> MIPS parts, 450MHz UltraSPARC IIs and the PowerPCs are all slower than
>+ >+ >> the 1GHz Pentium III for the SPECfp95 and SPECfp2000 tasks, and even
>+ >+ >> the 440MHz 8600 PA-RISC part in the N4000 is slower on SPECfp2000_base.
>+ >+ >
>+ >+ >I'll not argue with this, but note that the original poster said
>+ >+ >"... on similar MHZ CPUs ...". I wouldn't call 450 MHz "similar" to
>+ >+ >1 GHz at all: it's more than double!
>+ >+ >
>+ >+ >I don't know, but how does Intel @ 450 MHz compare to the other processors
>+ >+ >AT THE SAME CLOCK RATE? My guess would be similar integer performance,
>+ >+ >and worse FP performance.
>+ >+
>+ >+ The statement "RISC runs circles around Intel on floating point"
>+ >+ wasnt qualified. As for "the same clock rate", a quick scan of my
>+ >+ Dell catalog reveals no desktop systems available at speeds as
>+ >+ slow as 450MHz.. To me that says its a moot point to compare
>+ >+ at the same clock rate when Suns fastest processors dont clock
>+ >+ at as fast as Dells slowest desktops ($799 buys a fully configured
>+ >+ Dell 667MHz PIII system with a monitor and fast ethernet adapter,
>+ >+ compared to $5,500 for the 450MHz UltraSPARC II module (computer,
>+ >+ memory, disk, monitor and software sold seperately. Ack.))
>+ >
>+ >Yes, but try accessing over 4G of real memory on an Intel processor
>+ >with either Windows or Linux ( as it stands now ). You cannot.
>+ >And as for clock speed, what does that mean really? CPU speed is
>+ >not going to be your bottle-neck. I would be concerned more about
>+ >bus speed, in which yes, SPARC's run circles around most Intels.
>+ >SPARC's will run circles around Intel machines, they are designed
>+ >to.
>+
>+ Again, the STREAMS numbers:
>+
>+ (from http://www.cs.virginia.edu/stream/standard/Bandwidth.html)
>+
>+ Sun_Ultra60-360 1 355.2 343.8 311.1 358.6
>+ Intel_820_600B 1 379.2 386.4 448.5 448.4
>+ Intel_440BX_600 1 342.2 340.2 412.0 409.2
>
>
>Whoaa, those have gotta be outdated. You can get up to 1.8G memory
>throughput on the Ultra80.
A clue: the actual performance measured by actual code will be
less than theoretical, and the STREAMS benchmark is used to
measure actual usable bandwidth. As for theoretical, the Intel
840 chipset has a theoretical peak bandwidth of 3.2GB/s vs only
1.6GB/s for the 820 chipset shown above.
>+ The UPA bus on the Ultra60 at 360MHz is 120MHz, which I believe is
>+ the fastest interface on the workstations and workgroup servers.
>
>Nope SUN Ultra 80 with 1 450Mhz UltraSPARC-II CPU, which has
>4MB cache would be quite a bit faster. You can put 4G RAM
>in the machine, but with a 64 bit CPU, it can potentially
>address much more than that. The I/O interface can support
>up to 40MB per sec.
The UPA interconnect on the Ultra 80 x450s is 112.5MHz, which is
less than 120MHz. The Ultra 60 x360s have 4MB of cache per
chip as well. For tasks which run in cache, the Ultra 450 will
be quicker, but for tasks which are purely memory bound, the
balance shifts. For machines like the E450 with a 100MHz UPA
and only 400MHz processors, the balance shifts further.
>+ Given that it doesnt keep up with a mere 600MHz Pentium III on
>+ the slower 820 (which has half the theoretical bandwidth of the
>+ 840) I wouldnt be making blanket statements about bandwidth in
>+ all product catagories. Also, none of the workstations or the
>+ workgroup servers can take more than 4GB of real memory, so in
>+ this catagory, which represents the vast majority of the market
>+ (everything from $50,000 on down) your comments on addressing
>+ more than 4GB of RAM is moot in this segment.
>
>Moot? Really, you are comparing Sun workstations with Intel servers,
>that dont even (yet) support a 64-bit architecture.
When Im working with datasets that are larger than 4GB on the
desktop and could afford the paging on a 4GB machine, Id start
with a Compaq DS20 with a theoretical memory bandwidth of 5.2GB/s
and dramatically better integer and floating point performance,
or even an ES40 workstation with the same interconnect but with an
additional two processors and up to 32GB of RAM. Alternatively
the HP J series which top out at either 8GB or 16GB of RAM
(depending on the model) would be good candidates as well. 64-bit
addressability is really only a major win when you have more than
32-bits of stuff to address in the first place (and no, I dont
want to be paging in GBs at a time on a machine limited to 4GB.)
>Here are some points from Sun:
>
>Exceptional throughput.
>
>- UPA provides a crossbar-oriented interconnection establishing a
> 144-bit wide, ECC-protected data path to the CPU
>- Clocked at up to 112 MHz, the UPA crossbar gives a peak throughput of
> 1.8 GB per second (models with 450-MHz processor)
>- Memory subsystem offers a 576-bit-wide memory path
>- Architecture allows memory to be installed in fours to take advantage
> of 576-bit-wide memory path
>- UltraSCSI is integrated on the motherboard
>- A second UltraSCSI channel is available, allowing external devices to
> be connected to separately, further improving I/O throughput
All the same bits that my Ultra 60 has... No news here.
>+ When you want faster graphics than their custom 3D cards on
>+ the UPA bus, you can upgrade to the faster Sun Expert3D based
>+ on a PC graphics chip on the PCI bus, which (suprise suprise)
>+ is a PC based bus which is faster than Suns Sbus. Its about
>+ this time when someone usually jumps up to say Suns "flexible"
>+ plastic cases are of far higher quality than steel box PC cases,
>+ or that the same chips, disks and monitors which are used in PCs
>+ are of higher quality because they are shipped in a box with the
>+ Sun logo.
>
>So SUN uses PCI, so what?
With 3D chips developed and sold in the PC space first. The
"so what" part has to do with an unfounded response that by
not using PC parts something is automatically better, implying
that the systems which use them (PCs being their intended
target, and Suns top of the line workstations being their
unintended target) are somehow inferior. Blind statements
are also often made that Suns cases (cheap plastic that
flexes under the pressure of a pinky finger) are somehow
much better, and they hold that up as proof that Suns are
somehow more reliable.
>+ As for what the SPARC was designed for, even Suns management
>+ is on the record from a few years back as saying they havent
>+ been anywhere near the speed leader in a long time, but the
>+ software base of Solaris is what keeps them in the number one
>+ spot in the *NIX space. If performance and reliability were the
>+ only concerns, Alpha and OpenVMS would have eaten the *NIX market
>+ alive a very long time ago (does anyone remember the mid-1990s
>+ when DOS could say "ABORT, RETRY, IGNORE?" on a floppy read
>+ error, and yet Solaris 2.3 would reboot instantly without so
>+ much as a kernel panic?)
>
>Err, do you really think that Sun would be the number one in
>the UNIX world if they did not have good hardware to run Solaris
>on? I mean really, I work with a machine that has 64G of RAM in
>it and 53 CPU's.. Has Intel even came close to an architechture
>like that?
Sun sells hundreds of E10000s and hundreds of thousands of Ultra 5s,
10s, 60s and 80s every year. The bulk of their sales are on machines
which havent been performance comptative since the SPARCstation 2 was
overrun by the HP PA-RISC workstations. Just as the sales scale down
as the processor count went up, you can look at Intels largest system
(quantity 1?) which had 9,632 and as of June 7, 2000 still ranks #1
on http://www.top500.org's list of supercomputers. Suns highest
ranking system is #50, with only 720 processors. Note that different
tasks call for different underlying architectures, and that either of
these machines may be poor solutions for other tasks not measured in
their benchmarks.
>+ As Ive said before, Ive deployed Suns in a number of applications
>+ and I own a pretty substantial amount of Sun hardware and software.
>+ They provide "the right tool" for many jobs, but there are many
>+ more still for which they arent performance or price/performance
>+ competative with the alternatives.
>
>The difference is you are comparing Intel Servers, with Sun
>workstations. workstation != server, but even still, sun's
>workstations outperform the majority of Intel servers. CPU
>speed is only a small part of that.
You obviously havent looked at the Ultra 5S, Ultra 10S, or the
fact that the Ultra 80 is faster than the E450. Second, I
reference Sun "workstations and workgroup servers" multiple
times, and yet I do not use the word "server" with a PC anywhere
in the text you quoted. Finally, I know there is more to system
design that CPU speed, which is why I was able to consider and
approve the use of Sun hardware on my projects. In most cases,
software availability on scalable systems was the driving factor.
>Regards,
>
>anm
>--
>/*-------------------------------------------------------.
>| Andrew N. McGuire |
>| [EMAIL PROTECTED] |
>`-------------------------------------------------------*/
-Steve
------------------------------
From: "David .." <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup,redhat.general,redhat,alt.linux
Subject: Security
Date: Sun, 11 Jun 2000 23:40:44 -0500
http://securityportal.com/topnews/weekly/linux20000612.html
--
Registered with the Linux Counter. http://counter.li.org
ID # 123538
------------------------------
From: [EMAIL PROTECTED] (Jehsom)
Subject: Re: What should I do with ISO file?
Date: 12 Jun 2000 05:19:48 GMT
Paul <[EMAIL PROTECTED]> wrote:
> Hi! I download Redhat 6.2, but it is in iso files. How can I convert it to a normal
> files so that I can install it and also be able to copy it to a cd-r? Please explain
> more detail for me because I am a beginner. Thank you very much for your
> help. I am really appreciated.
You can burn it to a cd-r with your favorite cd burning program.
You can extract the files with winimage.
Moshe
--
jehsom@ angband.org, bellsouth.net, burdell.org, cc.gatech.edu,
nullity.dhs.org, polter.net, resnet.gatech.edu, wreck.org, yo.dhs.org;
gte741e, mj116 @prism.gatech.edu; jehsom, jacobsonconsulting @usa.net;
moshe@ medmanager.com; ICQ 1900670
------------------------------
From: [EMAIL PROTECTED] (Jehsom)
Crossposted-To: comp.os.linux
Subject: Re: Help - Questions...
Date: 12 Jun 2000 05:24:02 GMT
Kent A. Signorini <[EMAIL PROTECTED]> wrote:
> Some linux questions:
> 1) is there any way to fix those horrible fonts in Netscape? the net
> looks terrible
Standard problem with X and netscape. Next version of X and next version
of netscape fix this problem. Until then, scale your fonts bigger by set-
ting the font size on netscape settings.
> 3) i have two VFAT drives that I want all users to be able to wirite
> to--what
> setting do I make for their entries in FSTAB? i don't want to have to
> put all
> their UID's in fstab.
The manpage for 'mount' should say this. You should also check the man-
page for 'fstab'.
Moshe
--
jehsom@ angband.org, bellsouth.net, burdell.org, cc.gatech.edu,
nullity.dhs.org, polter.net, resnet.gatech.edu, wreck.org, yo.dhs.org;
gte741e, mj116 @prism.gatech.edu; jehsom, jacobsonconsulting @usa.net;
moshe@ medmanager.com; ICQ 1900670
------------------------------
From: [EMAIL PROTECTED] (Stephen E. Halpin)
Crossposted-To: comp.sys.sun.hardware,comp.sys.sun.misc
Subject: Re: Sun Sparc faster then intel pentium: is this true????
Date: Mon, 12 Jun 2000 05:33:23 GMT
On Sat, 10 Jun 2000 09:21:54 -0500, Dave Schanen <[EMAIL PROTECTED]> wrote:
>"Stephen E. Halpin" wrote:
>
>> You can go to http://www.spec.org and lookup the numbers.. Youll see
>> the dual processor 800MHz Dell topping out the quad processor 450MHz
>> Ultra 80 and the quad processor 400MHz E450 on SPECint_rate95 among
>> other things. Whereas the Suns are running at the highest clock rates
>> available today, the Dell can be had with faster processors now.
>
>This is getting getting ridiculous... Less than a month ago 'Boris' was
>in here quoting the same bullshit benchmarks you were. SPEC is based on
>theoretical compilers which the chip producers create, and don't reflect
>any real world technology, by the time we'll see the optimizations in
>Visual C++ these machines will be dinosaurs.
Clue: You can buy Intels tools TODAY through Intels web site. The C/C++
compiler (version 4.5, as used in the SPECxxx2000 tests) set are $299
list, and the VTune analysis package is $429. Add in Microsofts
Visual C++ environment for a list price of $549, and you have a full
suite of tools TODAY for $1,277, which is quite a bit cheaper than the
$1,995 you will have to pay to get Suns Forte C++ Personal Edition 6.
As for the SPEC numbers, there are reporting rules are quite clear
about not publishing numbers with research only compilers, and about
reporting availability, with the intent that anyone should be able
to buy the test bed as described and reproduce the benchmark results
within small tolerances. If anything, the greater difference in base
and non-base numbers on RISC parts is partly due to the allowance of
unsafe optimizations in the non-base tests, indicating that the RISC
vendors require more specialized, platform specific and unsafe
features for their compilers to produce fast code.
Its all there on Intels, Microsofts and SPECs web sites for you to read.
If you dispute the numbers, produce the results you come up with using
the tools and hardware listed in the SPEC reports. Otherwise, the only
"bullshit" is coming from you.
>Dave-o
-Steve
------------------------------
From: <[EMAIL PROTECTED]>
Subject: point to point
Date: Mon, 12 Jun 2000 05:30:11 GMT
I had 2 windoze machine connected point to point,I no longer use windoze
and use linux and wonder if anyone can point me in the right dirrection or
what to read to accomplish the same thing on linux a network with 2
machines.
Thanks
--
Posted via CNET Help.com
http://www.help.com/
------------------------------
** FOR YOUR REFERENCE **
The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:
Internet: [EMAIL PROTECTED]
You can send mail to the entire list (and comp.os.linux.misc) via:
Internet: [EMAIL PROTECTED]
Linux may be obtained via one of these FTP sites:
ftp.funet.fi pub/Linux
tsx-11.mit.edu pub/linux
sunsite.unc.edu pub/Linux
End of Linux-Misc Digest
******************************
